A Wizard whispers in your ear: "The password of Sheffield Library Packet Switching Service is ABC1234XYZ." That would be a conversation thirty years ago, on the Multi-User Dungeon, or MUD. The Dungeon was actually a minicomputer at the University of Essex. I won't tell you the name of the wizard, because he's a big noise in the …
Vax? What's wrong with it?
While Unix were constantly releasing new software/hardware to stay at the bleeding edge of technology, Vax was being stabilised.
Result? We have a VMS box which hasn't been switched off in 16 years.
No downtime, AT ALL.
It runs 24/7/365 and runs perfectly. Can't say that for any of our many Unix boxes. So before you use Vax to suggest old and decrepit technology, try bearing in mind that the old technology worked and still does.
If it ain't broke, don't fix it.
Local not-so-big iron
I remember where the local not-so-big iron was in Grimsby--I think it was owned by Imperial Foods, or leased by them or whatever.
The office block, last time I passed, seemed to be occupied by some sort of training company, and some of those seem to be only sustained by government-funded programmes for the unemployed.
Dollar worth three pounds?
In your dreams! I think you meant the pound was worth three dollars in them thar days or yore.
A pedant writes: I think the fixed exchange rate, in the days before currencies floated, was $2.40 to the pound. It had been $2.80 before the Wilson devaluation.
But that does not detract from a really interesting article.
I booted my Heath H11A a couple of days ago.
I have a small VAX cluster in storage (last fired up in 2006).
Sometimes I REALLY miss TOPS-10 and -20 ...
Addled thoughts from another dimension
"While Unix were constantly releasing new software/hardware to stay at the bleeding edge of technology, Vax was being stabilised."
"Unix" is not a company, but a an attempt at a cheap OS cobbled together at Bell Labs in 1969. It was a Doc Brown-style laboratory, complete with source code and was "ported" to various hardware. Since AT&T failed miserably in its attempts to sit on its "Intellectual Property" and since networking code was included in the BSD implementation, the laboratory never needed to be closed.
"VAX" is the DEC computer architecture, a hardware-software coevolution created by like Apple is famous for. Came complete with a wall of manuals that no-one ever read, and any access to the VAX hardware or OS code was guarded in-depth by operators, department beancounters and DEC T&Cs. Well, these days you can get OpenVMS I hear, as long as it's for "non-commercial" purposes and you register at HP or something. But who wants an OS that is handed out like a restricted firearm.
Dollar worth three pounds?
Indeed! I can remember when 2/6 was called "half a dollar" becaus a dollar was worth about five bob. It was a bit more than that in the seventies.
Oh, and RSX-11M is certainly the best operating system I've ever worked on and that includes VAX/VMS. You could do so much so efficiently (with the lovely Macro-11 assembler language) in such a small amount of memory.
And of course with PDP11s you could actually manipulate storage locations using the key switches on the front panel, so you could fix a bug in situ just by zapping a few instructions. Wonderful!
As AC said, VMS was quite somehting. Many years ago I inherited a VAX11/750 to manage along with one running Ultrix (remember that?) and a heap other Unix boxes. The best advice I ever got about running the VMS box was not to touch it. Utterly solid. Unlike the Unix boxes.
What is such a huge pity is that the son of VMS turned out to be anything but stable. (And I don't mean OpenVMS.) I think one lesson is that some ideas and designs simply don't scale.
The history of, and the long shadow cast over modern computer engineering by, the VAXen really is a story worthy of telling.
Understanding *why* they were stupid
"And the reason it was not launched was stupid: it was seen as a rival to the success of the 990 mini."
That's only stupid with hindsight. The traditional product lifecycle was spend some time developing the product and then sell it until it became obsolete. It is obvious *now* that in the IT business, products are obsolete even before the final stages of development and the only way to run the business is for products to be continuously under re-development and you sell them as soon as they are stable enough for the cost of customer support to be less than the profit on the sale. (That's why we have so much crap software.) However, even as late as the 1970s, that might not have been obvious to the senior management of any company.
Laugh at these poor souls if you will, (I did. I particularly enjoyed the suggestion that 95% of the computer's time was spent working out how much to charge for the remaining 5%. Makes modern OS schedulers look *very* efficient by comparison.) but do try to understand that the world really has changed since then. These weren't 21st century IT execs with under powered products. They were 19th century execs exploring a strange new business territory where the normal rules just didn't seem to work any more.
soul of a new machine
Ah, nostalgia. I think I'll go and boot my PDP 11/73, just for old times sake. It'll give me something to do while installing OpenSolaris on my PC...
I was at the Cambridge University Computer Lab at the time and we did get power cuts, but usually we got a warning phone call some minutes before so the IBM 370 could be stopped and the hard disks spun down. Not nice if it was halfway through your run ...
One of my lecturers once was involved in developing a compiler on Perkin Elmer hardware - not sure whether that was genuinely British or OEM'ed murrican hardware - circa 1983 I'd say.
Though I cut my teeth on HP minis, the VAXed were really very very good.
Need patriotic icons for long-gone computer systems.
>"A Wizard whispers in your ear: "The password of Sheffield Library Packet Switching Service is ABC1234XYZ.""
:-) I used to use a wide-open JANET node at UCL, no password needed at all: just dial-in,
PAD> call a040000496000001
and you were there. I forget where I got the phone number from, possibly one of my teachers at school told me, or maybe I just read it on a BBS somewhere. It was a whole load cheaper than playing Compunet MUD, but I liked the bugs on c'net MUD more.
Networks were fun then. You could wander around systems and they all had guest/news/games accounts to log in and give it a try. Once you'd found the numbers of the first few systems, you could browse around them and find directories with lists of other systems to log into. I used to play on CERN's mainframes now and again. It was really a case of EU-vs-US in those days though: as soon as you tried to get across the pond you'd run up against some kind of bitnet or timnet gateway that wouldn't let you through without an account, because over there it was all charged-for.
I still hold to the principle that the network is a public space and you have a right to wander anywhere that doesn't have an explicit "keep out" sign. An entire generation of coders grew up learning by exploring and sharing, and the skill base of the whole industry was the better for it. Now that everything's going proprietary, closed and IP/DRM-happy, how are people supposed to have that kind of 'apprenticeship' any more? And then we wonder why all MS' software is so shit.
(Hacker/pirate icon narrowly defeats smiley-face-nostalgia-for-the-old-days icon in a closely-fought battle!)
The first 16 bit microprocessor
Hmmm... I think the first 16 bit microprocessor was the Ferranti F100L, which I reckon must have been in silicon by 1974 'cos that's when I left Ferranti. It was a simple single-accumulator thing, and did arithmetic with a fast *serial* ALU (saved space and transistors).
vax/vms/alpha the best there was!
i "grew up" in my i.t. career working under vms, from 1.5 thru to 7.3
i mostly came thru data centre operations, thru senior operator, shift leader, then operations analyst (untold DCL routines i wrote)..and eventualy (sort of) a "systems analyst" (in the famous words of roy chubby brown "i dunno what it is loov, but i'll have a f*cking good look at it for ya!")..
as stated in the first post, vax/alpha/vms reliability was/IS the best. "5 9's" reliability (99.999%), with the alpha in 92 providing 64 bit computing and clustered hardware providing ultimate fault tolerance via soft switchable fail over..(..and this worked a treat, i saw it a few times).
M$'s "wolfpack" and "MSCS" , where are they now?!
personally, i think M$ had vms murdered, by allowing one of it's stooges (Compaq) to buy it up for $7bn in 97 and quietly suffocate vms, much to my eternal anger..i believe vms may have been sold, or it's still sitting in some dusty corner at HP central (the people who bought Compaq)..and, as seems to be the eternal case, the bigger the company, the worse it gets, so it's as good as dead at HP.
what ever became of the 64bit alpha chip i don't know, but i do remember there was a version of 64 bit windows that would run on an alpha, like 8 years ago..but that was quietly suffocated as well ^^
in this day and age of *requirements* for this type of simplicity of platform (ffs, how many versions of the IX's are there???) and UNTOLD reliabilty and ease of use (as anyone who ever worked on them will tell you, DCL is a *joy* to code, F$GETDVI anyone?!), why the hell wasn't vms the platform of choice for all critical apps?
(i don't know if this is still the case but apparently the NYSE still runs on vms, tho that could have changed it was a while since i heard that one..apparently they refused all overtures to get them onto something rubbish and i don't blame them!)
so, kids..before someone spouts off and tells you how "marvellous" any form of unix or linux is, remember, as far as server OS's go, you can't beat vms!
p.s. stufff and nonsense (but nothing about vms!): http://www.eupeople.net/forum
DEC: PDP-8 Family Computers
Minicomputers are truly fascinating, not least because they kicked-off the trends towards smaller, more affordable computers, but also because they set the scene for todays computers.
I'm fortunate, I have a microvax-II and a pdp-11 on which I ran the RT-11 (MSDOS-ish) operating system and Fig-Forth until the 36A PSU blew-up! It's amazing what people tolerated for computers and how much usefulness they squeezed out of such puny hardware. I love the way the processors are on hundreds of seemingly identical DIL chips; connected by thousands of wires strewn across multiple, enormous circuit boards. I love them so much I have a design for an original-speed Nova clone in roughly 10 DIL ICs; complete with a toggle-switch front panel. I'd build a Novella(tm) computer for anyone who was interested ;-) !
But DEC were certainly calling their crazy 12-bit PDP-8 machines computers as early as 1966 as the wikipedia article documents!
-cheers from julz @P
Oh the memories!
And what about ICL? Everyone in the UK wanted them there except the UK government who seemed not to care...
And yes, the VAX was superb to work on! As a consultant, the environments I worked in varied from legal, engineering, system support, satellite comms - it did it all and it worked. The command set (Digital Command Language, DCL) was consistent in its qualifiers - unlike unix and DOS. This made it easy to use and some of our users developed quite complex programs to aid their work. The only fault was that I could not have one at home!
And bear in mind that the PDP-11 (I worked on that too) celebrated its' 21st birthday since it was still being installed into body scanners in the 90s, it being simple, very fast and utterly reliable.
And the point of this article was ...?
Hey! Does this mean I'm now over the hill because I cut my teeth on RSTS on a PDP-11, then learning 6502 Assembler and Basic on an Ohio Scientific C3-A? Then SYS III on a DEC?
Sun Micro's 3/60 workstations while in school?
Or I am too young because I remember where I was when the Morris worm hit?
Yeah, I started young. Had my Cat 300 Acoustic modem then the 1200 "digital" (no hand set)
Remember all those home brew kits? Heathkit's H8, IMSAI, CP/M machines? Oh yeah. the memories....
Essex MUD ran on a DEC PDP-10, which wasn't a minicomputer - it was a mainframe, and sold as such.
I worked for Plessey (Poole) on the System X project for BT, and yes it had a lot of computing power as well as the specialised switching hardware. We tried to sell it overseas, but no-one wanted it. Huge amounts of the software dev was on Billing, so that trend wasn't unique to the mainframe bureaux.
Plessey would never have built a commercial computer product - they were into cost-plus contracts and they had absolutely no concept of managing product costs. It wasn't a company, it was an institution. GEC and STC (the other System X development partners) were even worse, we thought we were much more dynamic than them.
I designed a lot of the hardware and some of the software for the System X sub-system that linked the Operator Consoles into the network. It actually used microprocessors, and not just any old micro - it was Intel 8086 based! That turned out to be handy experience to have, and in 1983 I found myself in Birmingham working on PCs for Apricot. Good times, mostly.
Re:Dollar worth three pounds?
You obviously never bought computer equipment in the UK twenty to thirty years ago.
Those were the Days....
A DEC PDP 11/40 with 28KWords of Ram and a Single RK05 Disk Drive.
Run DOS V8 and get the odd F342-Odd Address or other Trap 4.
Guess what I used for my Degree Project?
The PDP 11 range was great but already showing signs of age when the Venerable VAX was Released in 1977. 32Bits - Brilliant.
You could have 20 people logged in and running applications on a 1Mb Machine. Can't do that today can we?
I remember fondly the 'pig'. This was the 240v->110v Inverter we had at Dec Arkwright Rd in Reading. The first VAX we had was a 110v beast.
Not in its Prime
Guy, I remember Prime computers. Napier College (now University) had one. You'd painstakingly type your program into the terminal, type "compile with pascal" (or "c with pascal" if you were feeling lazy) and then find your job at the end of a queue of 300 users. After a few pints and a couple of rounds of pool at the student union you'd come back to find that you'd missed out a semicolon in line 23 and the whole thing had failed.
The year I left they ditched the Prime and bought PCs with turbopascal.
we've come a long way...
and i'm proud to say that i have worked a fair bit on a PDP-11 (and later on DG AOS/VS systems) in the days of old, and have been in, and grown up with, the computer industry ever since..
my own first computer was a Sinclair ZX81. oh what fun you can have with simple things.
the DG 'minis' with the dumb terminals were the best fun of all (when you were a bureau operator that is)... proc on client's (remote) console anyone?? lol.
oh how things have changed.
i have to say that although modern computers and internet etc have made things far easier (on the user end at least), computers nowadays are definitely a lot less reliable (and trustworthy).
... so you see, i'm not just a troublesome spotty commentard after all..
DEC, HP, DG, etc minis still alive today!
If you have a fascination for how computing was in the 70s, or grew up with them as I did, you can get that retro experience with SIMH, a simulation platform for a variety of classic computers. I have both a VAX VMS and an HP2000 Access system running on my Mac (both OS are licensed for for free for hobbyist use). More info at http://simh.trailing-edge.com/
Incidentally, Essex ran MUD on a Dec System 10 and not a PDP-11. Oslo also had a copy IIRC. I recall BT PSS demo accounts were regularly traded for communications.
Personal as you like
Actually, early DEC computers *were* personal. Right back to the PDP-1, which very much resembles a primitive PC. A single box, with a single CRT screen and a single keyboard for input. One person sits in front of it and controls the whole machine. Programs are typed in, executed, and give results in real time. Quite unlike the mainframes that predominated in the 1960s. The PDP-1 has an excellent claim to be considered the first PC, except perhaps on grounds of price. It simply cost too much for anyone but an organization or a rich individual to acquire - but still vastly less than a mainframe. In fact, DEC always offered personal computers right up till the day it disappeared into Compaq. You could take a PDP8, or a PDP11, and use it with a single terminal. And the VAX and Alpha ranges incorporated workstation versions that were quite like today's PCs - only with much more stable (though far more expensive) software.
No mention of HP?
Interesting. A story by an author who I have always respected - Guy Kewney - and not a mention of one of today's big computer heavyweights that was to end up buying the remnants of DEC (in the form of Compaq, who took DEC over) - HP.
I worked for HP during the late 70's and thru the 80's. The HP1000 real time computer was a very definite competitor for the DEC PDP's. 16 bit architecture and all that, the F-series 21MX with its floating point processor plus a 256Kb memory store and 20Mb hard disk as big as an average washing machine supporting 5-10 users concurrently was a well configured system in those days. And here I am 20 years later wondering how I can make my 64bit CPU with 4Gb RAM and 2Tb of hard disk go faster. Oh, how the world changes.
Good work Guy, but I'm afraid on this occasion I can only sympathise in so far that you seem to have only covered half the story.
And Paris because....well just because.
But you haven't mentioned ...
But you haven't mentioned the other time sharing phenomenon, that of dialling up the provider in another city from a dumb terminal. I remember doing manufacturing process statistics by dialling a remote computer (on one of those new fangled Subscriber Trunk Dialling phone lines) and spending half an hour or more feeding preprepared paper tape into a dumb terminal, with the answers coming back to be printed on a device similar to a Creed teleprinter (but 7-hole tape not the 5-hole beloved of the telex system)! Despite the trunk call charges, and the bureau charges it was still said to be economic.
Oh yeah, the recession.
Really bad. 1970's - cooking a meal for 6 on a camping stove by candlelight while your Dad hung himself (by candlelight). These days it's not being able to take your second holiday abroad or decorate the guest bedroom in William Morris print wallpaper. Soft modern wanks.
I remember VMS at college
My abiding memory of it is that they told us when we started at college it was unhackable and 'military strength'.
To its credit, it did take me a week to break into it... in the process causing a crash of the kernel that apparently corrupted the main disk (that's what you get for only skim reading the manuals I guess).
It was a cool OS. If it worked on intel hardware I'd probably still have it on a VM somewhere. It's priviliege system is something that was only vaguley copied on NT later... and DCL has never been bettered.
Dollars and Pounds
Back then, the pound then was, I think, worth $2.40 in U.S. dollars then. But I think what was meant was that a U.S. dollar then was worth three of *today's* pounds, to help the British readers today make sense of the prices. But I suspect five or ten pounds would be closer.
TMS9900's and PDP11's
I can remember designing for the TMS9900 around 1980 - it was my first job out of university. I had to design both the hardware and the software - in those days, you were expected to be competent in both. The equipment we were designing was an industrial control system. Originally it was powered by a Texas 990 mini-computer, but we built out own CPU boards as soon as we could get hold of the 9900 chips - a big, expensive 64-pin chip in a white ceramic package.
They had a totally eccentric architecture, apparently deliberately designed to make them totally incompatible with everyone else's products. Once you started down the TI path, it was extremely difficult to switch to anything else. The address bus was numbered back-to-front (or was it the data bus - I can't remember now) and the peripheral chips would only work with the TI CPU's. They had the weird feature of keeping all the registers in RAM, apart from a single pointer register, so you could do a "context switch" (jumping to a different program), simply by changing that register. The theory was, apparently, that the speed of RAM was increasing so fast that the external RAM-based registers would soon be faster than internal ones. Later on we went to the cheaper TMS9981 and then the 9995, which was, I think, the end of the line for that range. The 9995 was a good processor, but it's non-standard architecture counted against it and it never really caught on.
Later, we designed a system which had two PDP11's working as intelligent disk drives (coupled to the infamous 300Mb disk packs) and a rack full of AMD bit-slice processors doing all the number-crunching - all controlled by a 6809! It was a data compression system using discrete-cosine transforms and was used by a big publishing house (Time-Life, I think), to send their pages by satellite to all the different printing presses. I think the FBI later bought one for storing fingerprints!
Happy days! Electronics isn't nearly such fun now.
Microfiche and birds on the wire
VMS came with (most) of the source code on microfiche, couple that with the Internals guides and you could really learn how the beast worked. And don't forget MicroVMS, it came on around 50 floppies of which at least one was going to be a dud.
Dialup over 300 baud could be exciting. The safest way to do anything was to create a command file and execute it. The alternative was to have noise on the wire complete commands for you with random additions. Sometimes those commands worked leaving you pale and wondering what you had just done as SYSTEM.
being too young to remember anything from these days, its a useful article on how the things actually happened. Thanks also to the commentaries for broadening the picture.
DEC (and ICL)
ICL was a creation of govt, and through its creation ICT's engineering discipline and innovation was frittered away. But in the 1970s I don't remember a shortage of PDP-11s - I was in effect the buyer of them in a University that I will not name because the central computing service boss thought that users getting their hands on computers would diminish his empire. It did.
@Destroy All Monsters
I don't need a lecture on Unix/VAX culled from a wikipedia page.
I'm well aware that Unix isn't a company. I was trying to keep it simple, for the sort of idiots who write terms such as "LOL".
My sincere pity if you were too thick or pedantic to understand my meaning.
About 10 years...
...after I started working on Burroughs mainframes a PC appeared on someones desk at work,
To this day I can remember thinking this means trouble..and in the days before IP networks were common you needed a different cable hanging out of the back of the PC to talk to each mainframe...I think they called it progress!
Just rebuild YET ANOTHER LINUX BOX - whilst reading this on the other screen, and it took me back. First job I ever had was with GEC, pursuading the boss that with all the money he spent buying time on the central mainframe, he could have his own PDP-11. Anarchy! The wrath of Weinstock fell heavily in those days and the arguement was long and hard. In the middle of this the DEC Salesman whispered "what about a VAX instead?" and so we got an early VAX 11/780.
What a simply wonderful machine? The power was awesome. A macro assembler that has never been beaten.
Just to let you know that MUD 2, the updated version of MUD - with 30 years of development is available to play for FREE at mudii.co.uk on port 23. This is the UK based version. There is also a version based in Canada at mud2.com. The Essex version of MUD, or British Legends as it is now know, is also available to play for free at british-legends.com.
You haven't lived till you've died in MUD!
I worked with VAX/VMS for about 20 years. A more reliable and secure OS you couldn't
wish for. Unlike what passes for an OS on the laptop I'm typing this out on at the moment.
I got out of the business when Unix and Windows took over. It was just too depressing.
Like going back to the stone age. At least VMS felt like it was actually designed and not just thrown together.
It wasn't ABC1234XYZ
It was NSS7HSD
"Isn't every computer a DIGITAL computer?"
Good grief, I remember that slogan. Whew. Them were the days....
What hurt me was buying a PDP11 (128K - yes, K!! words of ferrite core) and a couple of RL01 10 megabyte (megaword??) drives for 10 quid for the whole shebang, then selling it at a car boot sale for the same amount later. Reason? 'Cos when I span up the drives, the telly went on the blink...Missus NOT happy...
Article never mentioned the LSI-11, which was, like a microVAX, a PDP11 on a chipset.
Sod me, I thought the Intel 4004 was tricky....
In1966 working on a Monrobot XI (ah the joys of splicing paper tape) we certainly thought and spoke of it as a computer and relied on it to produce the stores orders, shipping lists, inventories and re-order prompts daily for our chain of supermarkets.
International Computers Limited
I'm really surprised that only one comment has mentioned ICL. The article talks of "...companies like Univac, Burroughs, NCR, Control Data Corporation, General Electric, RCA and Honeywell". And that, presumably, means ICL.
Perhaps I'm the only person reading this who named his son after an operating system. I suppose he could have been called TOS (Tape Operating System) or DOS (Disc Operating System) or Pick or CP/M but wasn't. He's called George after the best operating system of the lot---from ICL.
But this isn't the story of MUD and hackers (we can tell that tale another day). <--- I wanna hear this story.......
Anyone else get reminded of a BOFH episodes by some of the comments here :)
one word - simh.
I'm running a "Vax" and a "pdp11" in my shed right now, on a couple of old Linux boxes.
GEC attempted to make general purpose minicomputers, the 2000 and 4000 series. Default login was OPER/OPER.
Most of them were bought by GEC Marconi companies, who had to jump through numerous hoops to justify using a VAX rather than the much crappier internal product.
The reason there was no UK general purpose computer industry was largely the amount of government pork that was doled out to build largely useless weapon systems on a cost-plus (or quasi-cost-plus basis). This work was easier and more profitable than making things ordinary businesses wanted to buy.
I think this still continues (ID cards, nuclear power stations, Eurofighter, aircraft carriers).
%SYSBOOT-I-SYSBOOT Mapping the SYSDUMP.DMP on the System Disk
%SYSBOOT-I-SYSBOOT SYSDUMP.DMP on System Disk successfully mapped
%SYSBOOT-I-SYSBOOT Mapping PAGEFILE.SYS on the System Disk
%SYSBOOT-I-SYSBOOT SAVEDUMP parameter not set to protect the PAGEFILE.SYS
OpenVMS (TM) VAX Version V7.3 Major version id = 1 Minor version id = 0
%WBM-I-WBMINFO Write Bitmap has successfully completed initialization.
$! Copyright 2001 Compaq Computer Corporation.
%STDRV-I-STARTUP, OpenVMS startup begun at 30-NOV-2008 23:03:09.42
$ show default
$ show sys
OpenVMS V7.3 on node CI4IC4 30-NOV-2008 23:07:40.21 Uptime 0 00:04:44
Pid Process Name State Pri I/O CPU Page flts Pages
00000201 SWAPPER HIB 16 0 0 00:00:00.34 0 0
00000205 CONFIGURE HIB 8 5 0 00:00:00.50 112 175
00000206 LANACP HIB 12 50 0 00:00:02.62 331 1027
00000208 IPCACP HIB 10 6 0 00:00:00.10 99 177
00000209 ERRFMT HIB 8 22 0 00:00:00.12 130 211
0000020B OPCOM HIB 9 54 0 00:00:00.31 211 261
0000020C AUDIT_SERVER HIB 9 53 0 00:00:00.59 563 769
0000020D JOB_CONTROL HIB 10 27 0 00:00:00.16 191 342
0000020E SECURITY_SERVER HIB 10 38 0 00:00:01.30 1779 1687
0000020F TP_SERVER HIB 10 24 0 00:00:00.72 205 317
00000212 SYSTEM CUR 7 215 0 00:00:01.73 1463 356
This is running on a SIMH emulator presently running under NetBSD (-current) on an oldish HP i386 workstation. It is actually quite usable if one wants to remember the days of yore... I still haven't got a license for it, but one can login on the console and at least play with the system.
how about some specs
Ive looked at a lot of sites on the PDP-11 and I have a hard time figuring out what the specs were. The only thing Ive come up with is that it had 64k of RAM memory. What kind of permanent memory did it have? Just tape? Was the memory or OS accessed from a line editor, a teletype or console terminal? I know its hard to calculate processor speed as it had a lot of tiny processors, but what was its approximate speed? Thats really interesting to hear some of you have a working PDP-11 in your possession. And not to start a flame war, but which was better, Unix or VMS?
PDP-11 the first 'real' computer?
I'd guess the PDP-11 was the first 'real' computer -- bus based, extensible, accessible and so on. A nice bit of kit, made the PDP-8 look really ancient. I don't see much change since then, we seem to be stuck in an eternal timewarp where today's hot patents look like the stuff the mainframe builders were developing so they could cope with physically large computers for the target clock speed (same thing these days except everything's chip scale).
You should mention ICL and maybe its MICOS(?) processor -- a good seller, a microcoded processor capable of running different instruction sets. ICL was never going to make it into the microprocessor age, though -- can't think down to that level (or see that today's inadequate silicon is going to spaw tomorrow's version which will wipe the floor with you.)
The three day week wasn't about inflation. It was about coal. Prices were rising -- helped along by joining the EU, incidentally -- but wages were being held down by government order ("to stem inflation"). Various groups of workers didn't like it, including the miners. Rather than deal with them the government put the country on a three-day week to conserve fuel. I loved it -- three days is about the right length for a workweek, IMHO.